CN102034517A - Architecture and method for disk array to regenerate new data - Google Patents

Architecture and method for disk array to regenerate new data Download PDF

Info

Publication number
CN102034517A
CN102034517A CN2009101778160A CN200910177816A CN102034517A CN 102034517 A CN102034517 A CN 102034517A CN 2009101778160 A CN2009101778160 A CN 2009101778160A CN 200910177816 A CN200910177816 A CN 200910177816A CN 102034517 A CN102034517 A CN 102034517A
Authority
CN
China
Prior art keywords
data
disk array
volatility memorizer
hard disk
disk
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN2009101778160A
Other languages
Chinese (zh)
Inventor
陈重江
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Inventec Corp
Original Assignee
Inventec Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Inventec Corp filed Critical Inventec Corp
Priority to CN2009101778160A priority Critical patent/CN102034517A/en
Publication of CN102034517A publication Critical patent/CN102034517A/en
Pending legal-status Critical Current

Links

Images

Landscapes

  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The invention relates to an architecture and a method for a disk array to regenerate new data. The architecture comprises a disk array, a random access storage, a nonvolatile storage, and a hard disk, wherein the disk array comprises a plurality of hard disks with data; the random access storage is used for temporarily storing the data requiring reconstruction in the hard disks with data; the nonvolatile storage acquires newly generated data in the random access storage, wherein the data is generated after the operation of data of the hard disks; and the hard disk does not store any data and is used for receiving the newly generated data acquired by the nonvolatile storage and transferring the newly generated data to a corresponding section.

Description

The framework that regenerates new data and the method for disk array
Technical field
The present invention relates to a kind of framework that regenerates new data and method of disk array, especially refer to that a kind of new hard disk that disk array is provided regenerates the framework and the method for data, old disk array arithmetic expression is stored in the random access memory, and the operation result of the arithmetic expression in the random access memory is stored in the non-volatility memorizer, again the non-volatility memorizer operation result is write in the new hard disk, store the data of new hard disk by this non-volatility memorizer, can guarantee can be because of loss of power not produce the situation that data run off, so can reach new hard disk rise time of saving and system need not regenerate newly-generated data behind machine purpose.
Background technology
In one disc array system except RAID0, if the situation that certain hard disk (Hard Disk) is damaged has taken place, and when must more renew hard disk, the action of Rebuilding just must be arranged, and the meaning of Rebuilding will utilize the data of original hard disk to go to generate these data that have of new hard disk and be stored on this new hard disk exactly.
See also shown in Figure 1, it is the Organization Chart that existing disk array regenerates new data, wherein include two in the disk array and have data hard disk 11,12 and one does not store the hard disk 13 of any data, system then must will have data hard disk 11 earlier, 12 data are read in the volatile storage (can be a dynamic RAM (DRAM)) 2, carry out newly-generated operation of data according to the difference (RAID1/RAID5..) of redundant array of inexpensive disks 1 then, this operational data is deposited this hard disk that does not store any data 13, because hard-disk capacity is much larger than capacity, this action will be carried out in segmentation, carry out after the computing and write after the hard disk 13 that does not store any data, could finish a section (block) that regenerates at last, and regenerating progress (Progress) just can increase, and can be described as a loaded down with trivial details process.If in computing or write among the process of hard disk, the system failure has taken place causes shutdown or memory data to run off, so this to regenerate progress be uncompleted, and also must read again once again from original hard disk again, inconvenience very, and very consuming time and taking a lot of work, the present invention is just for solving the method for this problem.
Summary of the invention
Based on the disappearance that solves the above prior art, the present invention is a kind of framework that regenerates new data and method of disk array, fundamental purpose regenerates the framework and the method for data for the new hard disk that disk array is provided, old disk array arithmetic expression is stored in the random access memory, and the operation result of the arithmetic expression in the random access memory is stored in the non-volatility memorizer, again the non-volatility memorizer operation result is write in the new hard disk, store the data of new hard disk by this non-volatility memorizer, can guarantee can be because of loss of power not produce the situation that data run off, so can reach new hard disk rise time of saving and system need not regenerate newly-generated data behind machine purpose.
For reaching above-mentioned purpose, the present invention is a kind of framework that regenerates new data of disk array, and it includes:
One disk array includes a plurality of data hard disks that have;
One volatile storage, having in order to temporary those needs the data of rebuilding in data hard disk;
One non-volatility memorizer is obtained the newly-generated data after a plurality of hard disc datas in this random access memory carry out data operation; And
One does not store the hard disk of any data, receives the extremely corresponding section of data of the non-volatility memorizer of these newly-generated data.
For reaching above-mentioned purpose, the present invention is a kind of method that regenerates new data of disk array, and it is to include the following step:
The data that a plurality of needs with data hard disk in the disk array are rebuild copy among the random access memory;
To not need the data of computing directly to be stored in the non-volatility memorizer;
All hard disc datas in the storer are carried out XOR draw newly-generated data;
With the data storing after the computing among non-volatility memorizer;
To be stored in the data-moving of non-volatility memorizer to the section of the correspondence of new hard disk; And
Directly repay this section reconstruction and finish, can carry out duplicating of lower curtate data toward the upper strata.
Description of drawings
Fig. 1 regenerates the Organization Chart of new data for existing disk array;
Fig. 2 regenerates the Organization Chart of new data for disk array of the present invention;
Fig. 3 is the method flow diagram that regenerates new data of disk array of the present invention.
Description of reference numerals: 1,3~disk array; 11,12,31,32~have a data hard disk; 13,33~do not store the hard disk of any data; 2,4~volatile storage; 5~non-volatility memorizer; 61~data that a plurality of needs with data hard disk in the disk array are rebuild copy among the random access memory; 62~will not need the data of computing directly to be stored in the non-volatility memorizer; 63~all hard disc datas in the storer are carried out XOR draw newly-generated data; 64~with the data storing after the computing among non-volatility memorizer; 65~will be stored in the data-moving of non-volatility memorizer to the section of the correspondence of new hard disk; 66~directly repay this section reconstruction toward the upper strata finishes, and can carry out duplicating of lower curtate data.
Embodiment
Below in conjunction with accompanying drawing, be described in more detail with other technical characterictic and advantage the present invention is above-mentioned.
See also shown in Figure 2ly, for disk array of the present invention regenerates the Organization Chart of new data, it is to include: a disk array 3 includes a plurality of data hard disks 31,32 that have; One volatile storage 4, having in order to temporary these needs the data of rebuilding in data hard disk, and this volatile storage 4 is meant a dynamic RAM (DRAM); One non-volatility memorizer 5 is obtained the newly-generated data after these hard disc datas in this random access memory carry out data operation, and it is to utilize XOR (XOR) compute mode to obtain that these these hard disc datas carry out the data operation mode; One does not store the hard disk 33 of any data, receives the extremely corresponding section of data of the non-volatility memorizer 5 of these newly-generated data.
The nonvolatile storer 5 of above-mentioned indication, can be a volatile storer uses battery to provide electric power to preserve data, make it to become non-volatility memorizer, or utilize general memory card (for example: CF Card, SD Card), be even that inner flash memory (NAND Flash) carries out character stored in the calculating process, this kind storer or memory card capacity do not need too big, only bring to be used in store controller (Storage Controller) and to use and to get final product.
See also shown in Figure 3ly, be the method flow diagram that regenerates new data of disk array of the present invention, it includes the following step:
61~data that a plurality of needs with data hard disk in the disk array are rebuild copy among the random access memory, this step is that copying data that the needs of original disk are rebuild is among random access memory, carry out computing according to RAID kenel difference, and RAID1: can be directly in the time of the first step with data storing in non-volatility memorizer; RAID5: all hard disc datas in the storer are carried out XOR (XOR) computing draw newly-generated data.
62~will not need the data of computing directly to be stored in the non-volatility memorizer;
63~all hard disc datas in the storer are carried out XOR draw newly-generated data;
64~with the data storing after the computing among non-volatility memorizer;
65~the data-moving that will be stored in non-volatility memorizer is to the section of the correspondence of new hard disk, and this action is for instant input/go out requirement (I/O Request) not exert an influence; Aspect reading (Read): import/go out to read (I/O Read) and can't read (Read) action to new hard disk; In writing aspect (Write): write the data of new hard disk when writing (Write) if desired, then can write earlier among the non-volatility memorizer, be similar to general quick working area (Cache);
66~directly repay this section reconstruction toward the upper strata finishes, and can carry out duplicating of lower curtate data.
Put above-mentioned flow process in order, the method that regenerates new data of the present invention such as following:
1. the part that needs in original hard disk to rebuild is read out and be placed among the storer;
2. via calculating the data that be positioned over new hard disk, and be positioned among the non-volatility memorizer;
3. the neutral gear of handling again between the new hard disk I/O is positioned over new hard disk with data;
4. handle the step1 of execution again that in fact can also be synchronous after the step2 again and may not be certain to wait for that step3 finishes, so it is the progress of regenerating (Rebuilding Progress) is to come as foundation with the completeness of handling step2, more many than framework acceleration with method before.
5. the present invention and existing skill are compared, if advantage is that system unexpected mistake has taken place caused shutdown, only need reads calculating again at the uncompleted part of step2 when restart next time and get final product.
More than explanation is just illustrative for the purpose of the present invention, and nonrestrictive, those of ordinary skills understand; under the situation of the spirit and scope that do not break away from following claims and limited, can make many modifications, change; or equivalence, but all will fall within the scope of protection of the present invention.

Claims (6)

1. the framework that regenerates new data of a disk array is characterized in that it includes:
One disk array, it includes a plurality of data hard disks that have;
One volatile storage, in order to temporary described a plurality of have need the data of rebuilding in the data hard disk;
One non-volatility memorizer is obtained the newly-generated data after a plurality of hard disc datas in the described volatile storage carry out data operation; And
One does not store the hard disk of any data, receives the extremely corresponding section of data of the non-volatility memorizer of described newly-generated data.
2. the framework that regenerates new data of disk array as claimed in claim 1 is characterized in that, this volatile storage is meant a dynamic RAM.
3. the framework that regenerates new data of disk array as claimed in claim 1 is characterized in that, it is to utilize the XOR mode to obtain that a plurality of hard disc datas in this non-volatility memorizer carry out the data operation mode.
4. the framework that regenerates new data of disk array as claimed in claim 1 is characterized in that, this non-volatility memorizer is constituted by a volatile storage uses battery to provide electric power to preserve data.
5. the framework that regenerates new data of disk array as claimed in claim 1 is characterized in that, this non-volatility memorizer is by general memory card or inner flash memory is constituted.
6. the method that regenerates new data of a disk array is characterized in that, it is to include the following step:
(a) data that a plurality of needs with data hard disk in the disk array are rebuild copy among the random access memory;
(b) will not need the data of computing directly to be stored in the non-volatility memorizer;
(c) all hard disc datas in the storer are carried out XOR and draw newly-generated data;
(d) with the data storing after the computing among non-volatility memorizer;
(e) will be stored in the data-moving of non-volatility memorizer to the section of the correspondence of new hard disk; And
(f) directly repay this section reconstruction and finish, can carry out duplicating of lower curtate data toward the upper strata.
CN2009101778160A 2009-09-25 2009-09-25 Architecture and method for disk array to regenerate new data Pending CN102034517A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN2009101778160A CN102034517A (en) 2009-09-25 2009-09-25 Architecture and method for disk array to regenerate new data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2009101778160A CN102034517A (en) 2009-09-25 2009-09-25 Architecture and method for disk array to regenerate new data

Publications (1)

Publication Number Publication Date
CN102034517A true CN102034517A (en) 2011-04-27

Family

ID=43887293

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2009101778160A Pending CN102034517A (en) 2009-09-25 2009-09-25 Architecture and method for disk array to regenerate new data

Country Status (1)

Country Link
CN (1) CN102034517A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103699336A (en) * 2013-12-03 2014-04-02 中国科学院计算技术研究所 Method and system for distributing and reestablishing data of magnetic disc array

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103699336A (en) * 2013-12-03 2014-04-02 中国科学院计算技术研究所 Method and system for distributing and reestablishing data of magnetic disc array

Similar Documents

Publication Publication Date Title
TWI604308B (en) Data storage device and data maintenance method thereof
US8874833B1 (en) Sequential writes to flash memory
CN105988718B (en) Non-volatile memory device and control method performed by controller
CN100489808C (en) Storage system and bad storage device data maintenance method
JP2013539132A5 (en)
CN101604250A (en) The method of the control program of novel entities storage device more in the sas storage virtualization system
CN104461397A (en) Solid-state drive and read-write method thereof
EP2555106A2 (en) Method and apparatus of sanitizing storage device
CN104423888A (en) Data writing method, memory control circuit unit and memory storage device
US9785438B1 (en) Media cache cleaning based on workload
CN104699413A (en) Data management method, memorizer saving device and memorizer control circuit unit
CN104516959A (en) Method and device for managing database logs
US9639441B2 (en) Solid-state storage system, apparatus and method of writing data
TWI470432B (en) Electronic system and flash memory managing method thereof
CN102981969A (en) Method for deleting repeated data and solid hard disc thereof
CN102955751A (en) Storer storage device, storer controller and data write-in method
CN102890655A (en) Memory storage device, memory controller and valid data recognition method thereof
KR101548452B1 (en) Method and apparatus for recovering metadata in electronic device based on non-volatile memeory
CN103559119B (en) File operation requests disposal route and device
CN102508743B (en) Multi-point backup and real-time protection method for Linux file system
CN103914391A (en) Data reading method, memory controller and memory storage device
CN101661378B (en) Disk array 1 system and reading method for improving reading efficiency
US20090287882A1 (en) Raid_5 controller and accessing method with data stream distribution and aggregation operations based on the primitive data access block of storage devices
CN102737716B (en) Memorizer memory devices, Memory Controller and method for writing data
CN102034517A (en) Architecture and method for disk array to regenerate new data

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C02 Deemed withdrawal of patent application after publication (patent law 2001)
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20110427